MOREAUD, Stéphanie; GOGLIN, Brice; GOODELL, David; NAMYST, Raymond

doi:10.1109/IPDPSW.2010.5470849

Métadonnées

Licence d’utilisation du document

MOREAUD, Stéphanie
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Efficient runtime systems for parallel architectures [RUNTIME]

GOGLIN, Brice
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Efficient runtime systems for parallel architectures [RUNTIME]

GOODELL, David
Mathematics and Computer Science Division [ANL] [MCS]

Langue

Communication dans un congrès

Ce document a été publié dans

Workshop on Communication Architecture for Clusters, held in conjunction with IPDPS 2010, 2010-04-19, Atlanta. 2010-04p. 7 p.

Résumé en anglais

As the number of cores per node increases in modern clusters, intra-node communication efficiency becomes critical to application performance. We present a study of the traditional double-copy model in MPICH2 and a kernel-assisted single-copy strategy with KNEM on different shared-memory hosts with up to 96 cores. We show that KNEM suffers less from process placement on these complex architectures. It improves throughput up to a factor of 2 for large messages for both point-to-point and collective operations, and significantly improves NPB execution time. We detail when to switch from one strategy to the other depending on the communication pattern and we show that \ioat copy offload only appears to be an interesting solution for older architectures.< Réduire

Métadonnées

Partager cette publication !

Licence d’utilisation du document

Optimizing MPI Communication within large Multicore nodes with Kernel assistance

Langue

Ce document a été publié dans

Résumé en anglais

URI

DOI

Origine

Unités de recherche